Thomas Murray Response to CPSS-IOSCO Paper on Financial Market Infrastructure

Introduction

The events in global financial markets in the last four years have highlighted the importance of assuring that systemically important entities are sound and stable. Indeed, the crisis following the collapse of Lehman Brothers could have been far worse if key participants of the financial industry had defaulted as a result of the market developments. In this sense, the CPSS-IOSCO paper on “Principles for Financial Market Infrastructures” published in March 2011 (http://www.bis.org/publ/cpss94.pdf) is a prudent step to ensure that the industry is safeguarded against another potential future crisis, especially as financial markets are not yet stable and there continues to be a clear and present danger to market stability across the world.

As part of a consultative process, Thomas Murray has reviewed the CPSS-IOSCO paper and has prepared a two-section response. In the first part, we address some of the questions the committee asked in the cover note to the consultative report (http://www.bis.org/publ/cpss94covernote.pdf). The second part of the document will contain issues that we feel must be addressed or re-considered by the committee.

Part 1 - Questions Posted by the Committee

The cover note to the consultative report has asked questions on six specific topics: credit risk, liquidity risk, segregation and portability, general business risk, access and interoperability and finally, assessment of methodology.  The questions for the first two areas (credit and liquidity risk) will be answered in a single response. Admittedly, credit risk refers to counterparty risk and the principle is mainly addressed to CCPs, while liquidity risk has been given a broader scope covering all FMIs. Nonetheless, the questions in these areas are quite similar and to an extent can be addressed jointly.

1.1 Credit and Liquidity Risk

The committee asks what would be the pros and cons of establishing a minimum requirement that FMIs (but mainly CCPs) should make provisions in case of default of either “one” or “two” of its participants. Such provisions should assume that as a result of the default the FMI would face the largest exposure (in the case of credit risk) and largest potential open positions (in the case of liquidity risk) under extreme but plausible market conditions.

In our view, the question should be addressed from a different angle - we don’t see that the criteria be set to cover the default of one or two participants. We suggest that a different approach is preferable by looking at the market concentration. If a market is heavily concentrated then coverage for the largest two members should be sufficient to avoid a collapse of the FMI. However, such measures may be inadequate if the market has a moderate level of concentration. Take an example: assume that there are 15 participants in the market with relatively similar market share as shown in the table below (see overleaf).

Also assume that for a default scenario the FMI decides to adopt the “cover two” approach. This implies that the CCP would need to ensure that it has sufficient resources to cover the equivalent to 19% of all net obligations. However, if the situation is worse than expected and it includes the third largest participant, the FMI would not have resources to cover net exposures equivalent to 27% of total obligations.  Therefore, it appears to be a more adequate approach to decide on level of coverage (either for credit or liquidity) based on concentration levels. This risk should also cover systemic risk. If systemic risk is high in the market, then it is likely that the default of a single participant could lead to the default of two or more participants, therefore coverage would be insufficient.

Table 1

The thresholds of concentration aforementioned should be properly identified by the FMI during the period of stress testing. The basis of the stress test is defined by the “extreme but plausible market conditions”. This concept is frequently quoted by the committee but no guidance is provided on what parameters should be included when defining these scenarios.  If an FMI fails to accurately set the parameters for “extreme but plausible market conditions”, then thresholds are likely to be biased and thus, coverage for counterparty and liquidity risk would be inadequate. Admittedly, the degree of deviation of market conditions across markets makes it extremely difficult to clearly define what “extreme but plausible market conditions” means. Nonetheless, some effort should be made to shed more light on this issue.  For example, the committee could encourage FMIs to share their approach to stress-testing and their understanding of “extreme but plausible market conditions”, which would be at the core of their testing. Given the level of interconnectedness in the current global context, the exchange of such information would prove to be a helpful practice. Transparency in this issue may assist CCPs competing on risk.

1.2. General Business Risk

General business risk is covered by TM’s methodology for CSD under financial risk, which is the ability of the CSD to operate as a financially viable company. This risk concerns the financial strength of the depository and whether its capital is sufficient to meet the on-going operation of the organisation.

The committee is seeking comments on the pros and cons of establishing a quantitative and/or qualitative requirement for the amount of liquid net assets funded by equity that an FMI should hold to cover general business risk.  More specifically, the committee seeks advice on the convenience of establishing that amount to six, nine or twelve months of operating expenses.

Overall, the idea of ensuring that an FMI has resources that would allow it to fulfil its obligations during a number of months even if the market does not operate and revenues are not forthcoming makes perfect sense. The events in the Middle East and West Africa are clear examples of the relevance of FMIs having sound financial risk arrangements that would ensure the sustainability of the entity if the market closes for extended periods of time.

That said, the disadvantage of setting strict and/or prescriptive financial criteria is that it reduces the FMI’s room to manoeuvre during periods of economic hardship.  In this sense, the question of whether there should be enough reserves for a specific number of months is too prescriptive. There might be occasions during which a FMI might reduce its capital to cover certain liabilities and in doing so, might not be able to follow the proposed criteria.

In any case, the number of months of coverage should be secondary as there are more important issues that have not been considered in the document. Indeed, there are four items related to this issue that should be carefully considered before attempting to answer the committee’s questions. Such items are: First, the type of asset held as part of reserves. Second, the arrangements set in place to have access to those reserves. Third, the level of liability FMIs should accept. Fourth, additional resources an FMI may have access to such as bank guarantees and/or the explicit backing of the government.

Regarding the type of assets held as reserves, the key issue at stake is the quality of the assets that form the reserves. In some cases FMI managers may choose to invest in money market instruments and other short term securities in order to maximise the use of their reserves. The problem is that in the case of market breakdown (e.g. Egypt) or a major economic crisis (Greece-style), the value of the securities may plummet (dragging down the FMI’s level of reserves - even if such securities are government backed) or the FMI may not be able to liquidate  securities.

From a risk management point of view, perhaps the ideal approach would be to hold reserves in cash at the central bank. Of course, this is a suboptimal approach from a cash management perspective and the FMI’s boards will, therefore, continue to consider investment in short-term securities. In such cases, the committee could recommend that securities held as part of the FMI’s investment strategy comply with a minimum requirement in terms of quality; i.e. a risk rating threshold should be established. For instance, if the portfolio of the FMI includes public debt instruments, the sovereign risk rating of such country should be “investment grade” or above.

In addition, the committee should make some recommendations regarding the level of diversification of investment constituting the FMI’s reserves. More specifically, the CPSS-IOSCO document should set the ideal ratio of cash/securities that an FMI should hold as reserves, indicating the quality of those securities.

The next logical step is to decide where FMIs should hold their reserves. Of course, the securities side of reserves would be held either at a CSD or with the central bank (in case of government short term debt) if this is market practice. The key question is where should the cash side of reserves be held. Ideally, the option that minimises risk is to have FMIs deposit cash reserves with the central bank. This option eliminates the risk associated with holding reserves with a commercial bank. However, in some jurisdictions this might not be possible as only those organisations that have a banking license are allowed to have accounts with the central bank.

If it is extremely difficult to allow an FMI to have an account at the central bank and an FMI has no other option but to deposit its reserves with commercial banks, then the committee should consider issuing further guidelines for this area. Admittedly, the paper does indicate that the FMI’s management board should undertake a risk analysis of its custody and investment strategies. Nonetheless, the report fails to indicate that if reserves are to be held in a commercial bank, such institutions should meet a minimum level of risk rating and they must comply with the latest Basel capital requirements. Furthermore, FMIs should undertake due diligence when deciding which entities to use for banking. Also, FMIs must seek to distribute the reserves across several banks to minimise exposure to a single entity. More importantly, reserves (whether cash or securities) should be held at beneficial owner if market rules allow for the designation of securities and cash. This would ensure that if there is crisis, either at a CSD or custodian (in case of securities) or a bank (for cash), the FMI’s reserves are ring-fenced and protected.

The third element of key importance in general business risk is the issue of liability. Indeed, the level of liability that an FMI can accept is an essential piece of information in assessing the general business risk of the entity.  Of course, liability is often defined by local legislation and regulation. But the degree of divergence on the liability that FMIs face is quite significant. Taking the example of CSDs, in some markets the CSD will accept liability from direct losses if and only if these occur as a result of gross misconduct or the negligence of the CSD. In other countries, CSDs are not only liable for direct losses, but also for indirect or consequential losses.  This topic has not been discussed in the document and it would be ideal to have the committee’s view on the matter.

The fourth component to take into account is the additional support that an FMI may have during times of crisis. Access to other resources or the backing of the government (for instance in the case where an FMI is state-owned) makes a significant difference in the entity’s ability to survive during a prolonged crisis. While the paper does make reference to these aspects, it does so from a different perspective. The committee should amend its approach on the financial backing of an FMI to account for these other elements.

1.3. Assessment Methodology

Thomas Murray methodology for the risk rating assessment of CSDs provides an example of the areas that should be covered in order to establish the risk exposure of an FMI. The purpose of this rating is to determine the extent to which the depository minimises risk and maximises asset safety for participants and investors.  More specifically, our methodology establishes the risk exposures for investors in the post trade capital market infrastructure when transactions are settled while securities are held in a particular depository.  The rating assesses the effectiveness of a CSD and the processes used in post-exchange settlement and safekeeping to minimise investor risk exposures. CSD Ratings can be compared across markets. It should be noted that a CSD Rating does not assess trade execution risk but the settlement, safekeeping and asset servicing risks after trade execution.

For further information on this, click here.

Part 2 – Additional Comments and Suggestions

The first part of this document contained our answers to some of the specific questions posted by the committee in the cover note to the consultative report. In the following paragraphs we express our concern with some components of the CPSS-IOSCO paper.

2.1. Asset Servicing

The committee’s report does not properly address the relevance of asset servicing and asset safety.  As mentioned above, this is a key risk component that must be properly taken into account when assessing the risk exposure of an FMI such as a CSD.

Asset Servicing Risk is the risk that a participant may incur a loss arising from missed or inaccurate information provided by the depository, or from incorrectly executed instructions, in respect of corporate actions and proxy voting.

This risk arises when a participant places reliance on the information a depository provides or when the participant instructs the depository to carry out a corporate action on its behalf. If the depository fails either to provide the information or to carry out the instruction correctly then the participant may suffer a loss for which the depository may not accept liability. The depository may provide these services on a commercial basis, without statutory immunity, or it may provide the service as part of its statutory role, possibly with some level of protection from liability. This risk is likely to become much higher when international securities are included in the service.

Thomas Murray amended its CSD Risk Methodology in 2004 to account for the increase in asset servicing being taken on by CSDs. Potential losses from missed or incorrectly executed corporate events can be of significant magnitude, and few CSDs have adequate capital to protect  against this exposure (in comparison to banks for example). The committee may want to reflect on the ongoing debate in Europe in relation to the proposed CSD Regulation and the role of CSDs in non-core activities such as asset servicing.

2.2. Asset Safety and Account Structure

A key element that has been left outside the scope of analysis is the ideal account structure that CSDs should establish in order to minimise the risk faced by investor. There is an on-going debate regarding the benefits and disadvantages associated to either retail or whole sale models. The committee should take this opportunity to indicate its position in relation to beneficial owner accounts or omnibus accounts.

Clearly from a risk perspective, the beneficial owner account structure is preferable in giving protection to the end investor. However, the participants of a CSD, being the custodian or broker, typically prefer to have an omnibus account structure for operational efficiency purposes and will often put pressure on a CSD to adopt such arrangements especially where the CSD operates as a utility for the market participants. A clear statement from the committee on the desirability of the beneficial owner account structure to protect the end investor is therefore advisable.

2.3. Securities Lending and Short Selling

The committee has not made any comments or amendments to its position on the topic of securities lending; in fact, the recommendation on this matter dates back to November 2001 when the Recommendations for Securities Settlement Systems (RSSS) were published by CPSS-IOSCO.

The developments in global financial markets in the last four years - those same events that motivated the review of the principles for systemically important financial entities - have put securities lending; or more specifically short selling, in the spot-light. As a result of this, regulators across the world have taken reactive measures in this area. Indeed, there has been a clampdown on short selling (or various forms of this practice) in some European, Asian and American markets. The type of measures adopted by regulators has varied both in intensity and duration; in some countries there has been an outright ban on naked short selling. In other countries the regulators have decided to tighten short selling rules either on a permanent or temporary basis, but allowing this practice to continue, while in others it has required an improvement of reporting of short sales.

In some emerging economies, authorities’ attitude towards securities lending and short selling has been completely different to that adopted in developed countries. Of course, such a restrictive approach is related to the intrinsic features and characteristic of emerging markets. Admittedly, there have been a few changes in this area; indeed, in some Middle Eastern markets, authorities have relaxed their position on securities lending. Nonetheless, the committee’s silence on this particular topic is puzzling. If their view on securities lending as a whole has not changed, it is understandable that they might not amend their position. That said, CPSS-IOSCO should shed some light regarding their opinion on short selling in the context of the worldwide controversy surrounding this issue.

Equally puzzling is the committee’s silence on the current modus operandi of securities lending markets. The committee fails to address the need to ensure that securities lending is undertaken under a regulated framework that permits, not only the securities lending segment to operate efficiently, but also that ensures an optimal level of transparency. Much of the securities lending operates in the OTC market and given the desirability of moving securities transactions to the regulated market, it would have been expected that the committee would have provided some guidance in this area.

2.4  Use of Central Bank Money

While the use of central bank payment systems as vehicles to conduct the cash leg of settlement is mentioned in Principle 9, it seems that there is insufficient emphasis on the importance of assuring that all trade settlements are conducted through the central bank.

It is important that the principle be strengthened around this point as in some markets central banks have not been sufficiently progressive in developing their real-time gross settlement (RTGS) systems or have denied CSD or securities settlement systems (SSS) access to their payment system (such as in the case of Chile’s CSD).

If the use of central bank money cannot be utilised, then the payment system of commercial banks can be employed. However, the document should stress that the latter alternative should be considered a second best option as this practice does not efficiently minimise systemic liquidity and/or credit risk. If the financial stability of key participants is under pressure, it would be an assurance for investors, both domestic and foreign, to know that the cash settlement leg of all transactions is undertaken using the safest system available which is provided by the monetary authorities.

An issue related to central banks and that has not been addressed by the committee is the role monetary authorities should have in case a systemically important entity collapses. Should central banks have some kind of obligation to provide support to an FMI in case of default? If so, what kind of support should monetary authorities offer? More importantly, what changes in the regulatory or legal framework will have to take place in order to allow FMI’s to receive such support in times of crisis? This set of questions become even more important when considering the level of integration of financial markets across the world. Yet, the committee has failed to make a series of recommendations on this area.

2.5 Transparency

The report makes significant emphasis on the importance of transparency and the timely disclosure of information.  The spirit of the document is partially driven by the intention to enhance transparency in the industry. In this sense, the committee could make additional recommendations that would complement this general idea.  We have listed these ideas below:

• Principle 24 is dedicated to the disclosure of market data. However, the emphasis is made on trade repositories. The committee should extend the scope of this principle to all FMIs in order to promote a culture of market data disclosure and statistical analysis of market trends in the post-trade sector. The latter implies that the committee asks that FMIs engage more actively in data mining and analysis. At Thomas Murray we have noticed over the years that a large number of CSDs across the world struggle to provide basic information such as market concentration, data on failed trades and even on settlement activity.

• Disclosure of data should not be limited to direct participants. There are many other stakeholders that should have access to the information. Indeed, much of the data should be made publicly available unless there are strong confidentiality eg security reasons for not doing so; and even so, basic statistics should be disclosed openly through the FMI’s website.

• In those markets where English is not the first language, FMIs should disclose documents, rules and procedures in English. It is recognised that English is the common business language used across continents in the industry. Thus it only makes sense to publish all relevant information both in the market’s official language and in English.

• The committee discusses through the paper the relevance of conducting audits at different levels of an FMI. However, little is mentioned regarding the disclosure of the outcome of such exercises. The outcome of audits (especially operational audits) should be properly disclosed to the public to assure participants that the FMIs have taken the necessary steps to minimise, mitigate and prevent risk.  For example some CSDs have published public ratings and provided the outcome of SAS 70 type audits.

2.6 Exchange of Information and Further Clarification of Key Concepts

The committee has discussed in length the relevance of an FMI making adequate preparations for liquidity, credit and general business risks, as well as guidelines for collateral and margins. At the core of these principles is the use of stress tests, which are key tools that would enable FMI management and regulators to make optimal informed decisions regarding minimum levels of coverage, requirements, risk tolerance, etc. In addition, the committee clearly indicates that good stress-testing should account for “potential future exposures” and “extreme but plausible market conditions”.

In this context the recommendations of CPSS-IOSCO make perfect sense. The only drawback to this approach is that the committee fails to define what constitute “potential future exposures” and “extreme but plausible market conditions”.  While some work has been done in the past regarding stress-testing, the committee does not shed light on what constitutes these two concepts and leaves it up to the readers to understand this at their own accord.

The problem with this approach, as mentioned above, is that if the parameters that constitute “potential future exposures” and “extreme but plausible market conditions” are not properly defined, the outcome of all stress tests will be biased. Thus, the criteria and decisions made by FMI management and the regulator will be inadequate and the industry will be unnecessarily exposed to credit and liquidity risk.

The concerns regarding the definition of appropriate parameters for stress-testing are relevant for both developed economies as well as for emerging markets. The latest results of stress-testing in Europe were controversial in the context of the Greek debt crisis. In emerging markets the lack of experience that some FMIs may have is a paramount obstacle. Thus, without additional guidance the task may appear to be difficult.

Some claim that it is impossible for the committee to adopt a prescriptive approach to these concepts given the significant differences of markets across the world. Admittedly, it is far from ideal to define strict parameters for “potential future exposures” and “extreme but plausible market conditions” as country and market specific elements must be taken into account with special attention. However, there are ways to overcome this dilemma.

Our suggestion on this area is that the committee does not attempt to define its understanding of “potential future exposures” and “extreme but plausible market conditions”, and avoid a one–size-fits-all approach. We suggest that instead the committee encourages FMIs across the world to share and publish the parameters and conditions that apply according to their own conditions. While there are significant economic, political and legal differences across markets (no two markets are identical), there are also strong similarities that can be used to share experiences and expertise among FMIs. The exchange of information in this area would be extremely useful at three specific levels (i) it enhances FMI’s understanding of stress-testing and parameter definition; (ii) permits model calibration and (iii) slowly allows FMIs to take aligned positions in these areas, which could, in time, enhance the stability of markets across the world and allow FMIs to take coordinated actions in case of emergency.  The latter element is important given the interoperability and links between some FMIs.